Traffic Conflict Analysis of Non-Motorized Vehicle Waiting Areas at Signalized Intersection

In order to determine the impact of non-motorized vehicle waiting areas on the safety of signalized intersections, a random effect negative binomial model was established for typical three non-motorized vehicle waiting areas. The results show that the negative binomial distribution model with random parameters can make up for the defect that the traditional negative binomial model cannot describe the heterogeneity of the non-motorized vehicle waiting area, and the fitting result is better; the non-motorized vehicle waiting area in the form of BOX can effectively reduce the conflict; the volume of right-turning vehicles and straight-ahead no-motorized vehicles, the proportion of the large cars, and the proportion of the electric bicycles all have significant positive correlation with the traffic conflicts. The results are of great importance for improving the safety level of intersections.